Download FTS Fourier Transform Spectroscopy 2015S Interim Manual

yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

University of Toronto
Fourier Transform Spectroscopy
2015S Interim Manual
2015 January:
2014 September:
David Bailey <[email protected] >
David Bailey with contributions from students Dong Jin
Shin, Hazemm Daoud, Muhammadh Ali Abdullah
Salman, and Giuseppe Castiglione
Jason Harlow with suggestions from student Jerod
Wagman and references update by Barbara Chu
John Pitre
Please send any corrections, comments, or suggestions to the professor currently supervising this experiment,
the author of the most recent revision above, or the Advanced Physics Lab Coordinator.
Copyright © 2006-2015 University of Toronto
This work is licensed under the Creative Commons
Attribution-NonCommercial-ShareAlike 3.0 Unported License.
The Fourier Transform
The idea behind the Fourier transform is to represent a function f(x) defined in an interval 0<x<L as a
sum of waves:
f (x) = å an ei2 p nx/L
where the an are the amplitudes of the nth wave. This Figure shows the how a parabola can be better and
better approximated as more terms in the series are added.
Figure 1: Partial Fourier sums for f (x) = x2
In the continuous limit, a Fourier series becomes a Fourier Transform. Two functions defined by
f (x) =
ò g(s)e
i2 p sx
ò f (x)e
g(s) =
-i2 p sx
are said to constitute a Fourier transform pair. f(x) and g(s) are called Fourier transforms of each other.
If f(x) and g(s) are even functions then the imaginary parts of (1) and (2) disappear and thus
f (x) =
ò g(s)cos(2p sx)ds
g(s) =
ò f (x)cos(2p sx)dx
Equations (3) and (4) are called cosine Fourier transforms. Cosine Fourier transforms are more
commonly introduced in the following way.
If f(x) is defined only for 0 < x < ∞ then f(x) can be represented by
f (x) = 2 ò g(s)cos(2p sx)ds
g(s) = 2 ò f (x)cos(2p sx)dx
Equations 5 and 6 constitute a cosine Fourier transform pair. An analogous set of equations define the
sine Fourier transform.
The Fourier Transform and the Michelson Interferometer
Consider a beam of light incident on the beam splitter with an electric field given by
E(x, k) = E0 (k)e (
i kx-wt )
where k=2π/λ is the wave number. The flux density B(k) is proportional to E0 (k). In this derivation we
will ignore all proportionality constants since we will eventually only be interested in the variation in a
signal and not in its absolute magnitude.
In the Michelson interferometer (Figure 1), after amplitude splitting there are two beams which have
traveled distances x1 and x2. They recombine to form a resultant electric field ER(x1, x2, σ). Each beam
has undergone one reflection from, and one transmission through the beam splitter so both should have
the same amplitude. The resultant electric field, aside from a proportionality factor to account for losses
E(x1, x2, k) = E0 (k)[ei(kx1-wt) + ei(kx2 -wt) ]
Figure 2: Schematic of a Michelson Interferometer
The intensity (e.g. in W/m2) for a particular value of k is given (within a multiplying factor) by the
square of the electric field:
I R = ER* ER
which, written explicitly, becomes
I R (x1, x2, k) = E02 (k)[2 + eik(x1-x2 ) + e-ik( x1-x2 ) ]
I R (x1, x2, k) = 2E02 (k)[1+ cos(kd )]
where δ=x1−x2. The total flux for any path difference δ is obtained by integrating over k:
I R (d ) =
ò I(d, k)dk
I R (d ) = 2 ò E02 (k)dk + 2 ò E02 (k)cos(kd )dk
Equation 13, which gives IR(δ) will have a constant part and a varying part. The constant part could be
determined by measuring IR(0) when the path difference is zero or by measuring IR(∞) which is the
constant level to which the signal tends when the path difference becomes very large (see Bell p.41 for
further information on this point). In actual fact, if the beam splitter in the interferometer does not split
the incoming beam into two equal components then there will be a constant flux, I, which is independent
of the path difference and this will add another constant term. If we agree to interpret IR(δ) as the
fluctuation of the interferogram from its final constant level and not from zero then equation 13 becomes
I R (d ) = 2 ò E02 (k)cos(kd )dk
Again, ignoring proportionality constant we may replace 2E02(k) by B(k):
I R (d ) =
ò B(k)cos(kd )dk
But this is just one of a cosine Fourier transform pair and so we may also write:
B(k) =
(d )cos(kd )dd
So within a multiplicative constant the incident power spectrum can be computed by a cosine Fourier
transform of the interferogram.
Safety Reminders
 Do not look directly into the laser beam. This laser is not intense enough to require eye protection,
but forcing oneself to stare into any laser beam could cause eye damage.
 DO NOT TOUCH ANY OPTICAL SURFACES! This is not dangerous to you, but fingerprints on
the surface of mirrors, lenses, beam-splitters, … will degrade your signals and damage the
NOTE: This is not a complete list of every hazard you may encounter. We cannot warn against all
possible creative stupidities, e.g. juggling cryostats. Experimenters must use common sense to assess
and avoid risks, e.g. never open plugged-in electrical equipment, watch for sharp edges, don’t lift tooheavy objects, …. If you are unsure whether something is safe, ask the supervising professor, the lab
technologist, or the lab coordinator. When in doubt, ask! If an accident or incident happens, you must
let us know. More safety information is available at
Michelson Interferometer
Figure 2: A possible schematic of the experimental apparatus.
For device specifications, search for the model numbers at, or ask the Lab
Technologist for the device manuals.
HeNe Laser
Piezo Controller
Source of red laser light, with a well
defined peak at 630 nm
Used to calibrate data and correct noise
from non-constant stepper velocity
Allows extremely fine control of orientation
and position of fixed mirror
White light source includes all wavelengths
in visible spectrum
Connected to setup with fibre optic cable
Focused through samples to perform optical
OSL1 White Light Source
Silicon Detector
Measures intensity of light incident on
This experiment contains two: one to acquire
inteferograms (with white light); the second
to detect the laser light to correct for
deviations from non-constant stepper motor.
Translation Stage + Actuator
Responsible for holding and controlling
motion of movable mirror
Motion can be controlled with provided
software, or manually with onboard
Stepper motor may have non-constant
velocity, which can complicate time-domain
measurements. Data Analysis software
employs algorithms to correct this deviation
based on HeNe callibration data.
Translation Stage + Actuator
Used to manually adjust position of movable
Relay lenses, beam splitters, and mirrors
Steer beams to determine optical path
It is essential that the laser light be aligned
properly, otherwise fringes cannot be
Placed in sample holder to filter white light
source in initial experimental investigations.
Note that the transmission characteristics of
interference filters change if the incident
light is not parallel. The wavelength of
maximum transmission shifts to lower
wavelengths and the transmission curve
broadens. You will be measuring the
transmission characteristics of filters and
comparing them to measurements made
using a monochromator, so it is important
that the light be parallel.
Green filter is recommended for finding the
zero point, since the light passed is the
brightest among the filters.
Replaced by other samples, e.g. ink in water,
for optional investigations.
Interference filters
Find a rectangular piece of stiff white piece of paper and a plastic ruler.
Switch on the HeNe laser and turn off unnecessary lighting.
Place a rectangular piece of stiff white paper flush with the front of the laser and mark the beam
spot. The paper can now be used to check that the beam is horizontal throughout the set-up. the
middle picture below is good (the mark is invisible under the laser spot). The third picture shows
a problem since there are two spots, neither at the right height; a mirror needs adjustment.
A transparent ruler can be used to check alignment of reflected
beams; the picture to the right shows a reflected beam that is
not aligned to the initial beam.
Place a white screen (e.g. a white piece of paper or ground
glass) at the location of the detector.
If there is no apparent interference pattern, check the position
of each output light through the interferometer by blocking
each light path.
Adjust the interferometer mirrors using the knobs on their
backs such that the position for each light output is the same.
From here, adjust mirrors carefully using the adjustment knobs
until you observe interference fringes
Adjust the mirror screws and move until getting the widest
possible circular interference patterns. (Tip. Move the screws
in the direction that it gives you wider spacing between two consecutive fringes. As you do this,
the fringes become more and more curved. Since the central fringe is bigger than the beam
splitter cube in use, if aligned perfectly, you should be able to see rectangular uniform intensity
1. Since seeing the circular fringes is difficult when the distances of the two arms are equal,
move the actuator so that the difference in this distance is about 1cm. (The field of view of
the interference pattern should increase as the distance diff between the arms increase. Why?
This means that you should be able to see more fringes per area as the distance difference
2. Now use the alignment tip mentioned above to make sure that the interference pattern is
circular. As you do this, make sure that the circular interference pattern is centered with
respect to the shadow of the rectangular beam splitter cube.
3. To help you with very fine alignment, turn on the piezo controller and use the X,Y,Z knobs
to center the circular interference pattern with respect to the shadow of the cube.
Figure . Location of the piezoelectric crystals on the fixed mirror.
4. At the end, you should be able to see many concentric circular fringes that is the
characteristic of a Michelson Interferometer. (Why do you see circular fringes ?)
5. Basically, you are trying to do the same thing when the difference in the distance of the two
arms of the interferometer is zero. Except it would be harder to get the feel of alignment
since the central circular fringe will take up the entire field of view.
Set the arm distance such that the difference is zero. Use the same approach to alignment as
Once everything is aligned, take away the sheet of paper that you used to make way for the light.
Data Acquisition
Use the ”Signal Express” program
1. Click Add Step
→ Acquire Signal →DAQmx Acquire → Analog input → Voltage
2. Choose the aio and ai1 channels
3. Set Step Setup:Timing Settings:Acquisition Mode to Continuous Samples
4. Choose your sample Rate and number of Samples to read.
5. Press on the DataView tab
6. Click Run
7. Expand DAQmx Acquire
8. Setting up Digital Filter in Data Acquisition:
9. Drag Dev ai0 channel into the window channel. You should now be able to observe some signal.
10. In order to isolate the noise software based filters can be applied to the signals: Press on Add
Signal Processing
Analog Signals Filters
11. Customize the filter specifications through the tool box. The following filters are recommended:
(a) For the ai0 channel (Detector Type: DET36A): (This filter is recommended for the HeNe
light source)
i. Specify input data for the filter: ai0
ii. Type: IIR Butterworth Bandstop
Filter iii. Upper and Lower cutoffs based on the frequency of noise. E.g. 60Hz noise
would need a bandstop of 50-70Hz
(b) For the ai1 channel (Detector Type: PDA36A): (This filter is recommended for the
Broadband light source)
i. Specify input data for the filter: ai1
ii. Type: IIR
Butterworth Lowpass Filter iii. Cutoff: 200 Hz
12. Add two new displays and drag the two filtered data outputs to the displays.
13. Acquiring data: To record the data, click on the Record button next to the Add Step button. This
opens up a pop-up window where the data channels to be recorded can be selected, as well as the
name of the record and an optional description.
Moving Mirror Motor Control
Double click on APT User1 program on computer Desktop. If it is not there, go to: Start Menu →
All Programs → Thorlabs → APT → APT User.
 Set the Jogging to Single Step, and choose the Max. Velocity and Step Distance; 0.05 mm/s and
0.001 mm are reasonable initial values.
 Operating
1. Click the position display window, a pop up will show up that asks for the position in millimeters
you want the actuator to be in.
2. Specify the position and press Enter. This should move the actuator with the velocity specified in
the settings.
3. (Warning: if you try to move to a position that exceeds the physical limits of the actuator, the
software will give you warnings.)
Experimental Procedure
HeNe Laser
1. Align the setup for the HeNe light. Make sure the arm distance difference is zero.
2. Use the Data Acquisition procedures to set up data acquisition for one of the detectors. For the
HeNe part of the experiment, you only need one detector.
3. Make sure the output of the interferometer hits one of the detectors. An iris or home-made
aluminum foil cover with a pin-hole may be used later reduce noise or improve resolution.
4. Before recording the actual data, set up the motor control using the Motor Control Procedures.
5. Set the actuator to move about 1mm. (If your alignment is not excellent, moving the actuator too
far may move your system out of alignment.) Do not move the detector just yet.
6. Record the data and while it is recording, quickly go back to the actuator control and move the
actuator by pressing Enter.
7. After the stop recording the data when the actuator is stopped.
8. The recorded data can be found in the Logs section in the Signal Express software. Right click
the individual data streams (usually named filtered data, filtered data1) to open their fi location in
Windows. The file can be opened only with Excel, so remember to save them as .xlsx or any
compatible format if planning on analyzing the data later. Isolate the points corresponding to the
interference pattern and create two files with the data points from both detectors. Note: It is
important that there is a one-to-one correspondence between the HeNe data and the data for the
light under examination in order for the phase correction software to work properly.
9. Make a FFT of the data to assess its quality.
White Light
Finding the zero point
 To observe interference fringes for broadband light such as the white light, you need to
set the two arm length distance to be exactly equal within few micron of tolerance. (Why?)
To find the position of the actuator where the broadband interference fringes can be
observed, first carefully measure the distance of each arm length using a plastic ruler.
After recording the positions, move the actuator so that the arm distances are roughly the
Using the HeNe light only, adjust the setup at this position using the Alignment
procedures. The alignment for the HeNe light should also align the setup for the broadband
light source.
Turn off the HeNe light and turn on the Broadband light source. To see the interference
fringes for this light source, use the J-Jog feature of the actuator controller, or the
controller software and move the actuator around the expected zero difference position.
As you do this, you should be able to see interference fringes for the broadband light
source if you put a piece of white paper like you did in the alignment proce- dure section.
Now, use the Alignment Procedures again to align for the broadband light this time. Use
the procedure with the broadband light only.
Make sure the alignment is good across all positions where the interference can be observed.
(Note, the position where the interference fringes are the brightest is where the difference
in the two arm of the interferometer is exactly zero. However this is not always true, why ?).
Detector setup
Turn on the HeNe laser light and the Broadband light.
The output of the interferometer should split into two beams of light at the last beamsplitter
cube after the focal lens.
For the HeNe light detection, put the detector on the right side of the beamsplitter cube such
that the ONLY HeNe light dot hits the detector.
For the Broadband light, put the detector on the bottom of the beamsplitter cube such that
only the broadband light dot hits the detector.
This ensures that each detector reads light from individual light source independently.
(Note: If the interference signal is weak when you take data, try narrowing the light dots by
moving the detectors backwards and forwards. This should focus more light into the
detector for improved intensity)
Running and measuring data
About 0.5mm away from your measured position where the arm length diff is zero.
Go to the actuator controller software, set the actuator to move about 1mm. This should scan
all range of the interferogram. However, do not move the actuator just yet.
Use the procedures for the data acquisition to acquire data for both detectors (Note, record
the data that is filtered using the recommended digital filters.)
Press record and afterwards quickly move the actuator to your specified position.
Once the actuator stops moving, stop recording the data.
In the acquired data, you should crop the regions where there is actual interference for both
detector signals (Why? Hint: Alignment). (When you crop the data, make sure you crop
both HeNe and broadband light data at the same places in time, and make sure that they
have the same number of data points)
Analyze the interferograms and obtain the spectrum of the light under examination.
You may use a variety of programs to analyze the data. Sample Python analysis code is provided (but as
of 11 September 2014 it has not been debugged). The code is quite involved. Do not use the code as a
black-box. An essential step in understanding the experiment is to go through the code and identify the
analysis process.
This code uses data provided by the user for the intensity of the light being examined and data for the
HeNe laser taken at the same time by the two detectors. Due to the non- constant speed of the steppermotor, and the need for the data to be taken at constant phase intervals, one cannot just Fourier
transform the obtained data without modification.
With this in mind, the code does two things: a) It accounts for the non-constant velocity by comparing
the data of the examined light with the HeNe data and then creates a vector of new data points
corresponding to points at constant phase intervals. b) It performs a Fast Fourier Transform
As the data is taken in time domain and the motor moves at a non-constant speed the data for the HeNe
laser deviates from a sinusoidal by being narrower or wider between different zero crossings. By
assuming a phase of pi between consecutive zero crossings one can assign phases to the data obtained
and in this way one can then take data points corresponding to constant phase intervals.
One then performs a discrete Fourier-transform on the data to obtain the power spectra of the examined
A more detailed description of the code can be found in appendix III. To run your analysis, simply
specify the file path to your data in the provided code.
1. Before beginning the experiment, trace out the path the light will follow on the bench.
2. Explain why changing the path difference does not change the brightness of the fringes for the
HeNe laser, but it does for non-coherent sources (e.g. white/filtered white light).
3. The fringes in the Michelson interferometer are called fringes of equal inclination. What does
the term “equal inclination” refer to?
4. Consider the situation where the mirrors are exactly parallel. If the lens and detector were
removed from the output arm of the interferometer and replaced by a white screen, would
interference fringes be seen on the screen? Why or why not? Does your answer depend on
whether or not the separation between the mirrors is zero?
5. As the distance between the mirrors becomes smaller what happens to the ring pattern? Why?
6. The Convolution theorem states that given a function h(τ) = (f(t)g(t-τ)dx (the convolution of f and
g), then its Fourier transform is given by h(w ) = f (w )g(w ) (where h, f , g are the Fourier
transforms of h, f, g). This can be used to show that the Fourier Transform of Γ11(τ) is the power
spectrum. The correlation function for two electric fields, Ei and Ej is defined as Γij(τ) =
Ei(t)Ei*(t+τ), where the electric fields are averaged over t. In the case of a perfectly
monochromatic beam with E1(t) = E0 e−iω0t, the autocorrelation function is Γ11(τ) = E1(t)E1*(t-τ)
= |E0|2 eiω0τ. What is the power spectrum P(ω)? Is this what you would expect? Why?
7. A more realistic coherence function is given by Γ11(t) = |E0|2 e−iω0τ e−l0τ/c, where l0 is the
coherence length of the source. The resulting Power Spectrum is a Lorentzian centered on ω =
ω0, with FWHM l0/c. Use this to compute the coherence length of the red light.
8. [Answer this question if you have used the red (632.8 nm) filter with the narrow (3.6 nm)
bandwidth.] The transmission curve for this filter is approximately Gaussian. For a Gaussian,
I(δ) is given by
æ p ö2 æ Dl ö æ d ö
-ç ÷ ç ÷ ç ÷
è 2 ø è l0 ø è l0 ø
I(d ) = ke
where k is a constant, Δλ is the full width at 1/e points in the transmission curve, and λ0 is the
central wavelength of the transmission curve. When I(δ) falls to 1/e of its maximum at δ= δe
then from the equation above,
0  ( e /  )
Find δe in units of λ0 from your interferogram and calculate Δλ by assuming a value for λ0. Does
this value agree with your FT data?
1. Bell, R.J. (1972). Introductory Fourier Transform Spectroscopy. New York: Academic Press
2. Brigham, E.O. (1988). The Fast Fourier Transform and its Applications. Englewood Cliffs, NJ:
Prentice Hall.
3. Fowles, G.R. (1975). Introduction to Modern Optics. 2nd ed. New York: Holt, Rinehart and
Winston. (Also Dover Publications, 1989).
4. Hariharan, P. (2003). Optical Interferometry. 2nd ed. Amsterdam; Boston: Academic Press.
5. Hecht, E. (2001) Optics. Addison-Wesley 4th ed.
6. James, J.F. (2002). A Student’s Guide to Fourier Transforms: with Applications in Physics and
Engineering. 2nd ed. Cambridge: Cambridge University Press.
7. Jenkins, F.A. and White, H.E. (1976). Fundamentals of Optics. 4th ed. New York: McGrawHill.
8. Smith, S. W. (2002). The Scientists and Engineer’s Guide to Digital Signal Processing.
9. Stein, E.M. and Shakarchi, R. (2003). Fourier Analysis: An Introduction. Princeton, NJ:
Princeton University Press.
Figure I.1: Spectrum obtained for white light passed through a red filter